College Athletes Under Pressure: Eating Disorders Among Female Track and Field Athletes

Abstract

Eating disorders are a widespread problem for college female athletes. This article builds on this research by collecting survey data from female college track and field athletes at 30 universities. Results find that the division level is not a significant driver of eating disorders. Another significant finding comes from breaking out the pressures into its team-related and social causes. Results find that team-related pressures lead to more vomiting behavior, while social pressures impact dieting. Previous studies which aggregate pressures, rely on one eating disorder, or focus on a single university for their sampling may not reflect these complex relationships. Even after controlling for individual factors, competition level, and team nutritionists, the pressures still impact these athletes. This suggests a need for additional screenings and counselors to address pressures occurring both on and off the track. While difficult to achieve, there needs to be a change in team cultures across all divisions.

JEL Classifications: I1, Z2

Keywords

eating disorders college athletes pressures

Introduction

Eating disorders are a major public health crisis in the United States, affecting more than 30 million people in the United States (Hudson et al., 2007; Le Grange et al., 2012). With the median onset age for eating disorders being approximately 18 years old, it has become a particularly widespread problem on college campuses (Volpe et al., 2016). Eating disorders are more heavily skewed toward women, with a 3:1 female to male ratio among college students (Eisenberg et al., 2011). Among college students, female athletes are particularly susceptible to developing eating disorders. This is even more likely to occur among runners and other athletes in sports which emphasize leanness (National Center on Addiction and Substance Abuse [CASA], 2003). This is concerning as eating disorders have been shown to cause poor physical and mental health. Eating disorders have the highest mortality rate among all mental illnesses (Smink et al., 2012). And not only does experiencing eating disorders lower the quality of life for young people, it also puts a strain on the entire family (Gilbert et al., 2000; Muñoz et al., 2009).

Eating disorders include a variety of psychological disorders that have to do with irregular or disturbed eating behaviors. There are various signals of an eating disorder which include obsessions with food, body weight, and shape. Two of the most common eating disorders are anorexia and bulimia. Anorexia is characterized by self-starvation through excessive dieting causing dangerous levels of weight loss. People suffering from bulimia engage in binge-eating followed by vomiting. This bulimic behavior can be a repeating cycle. While anorexia and bulimia are different behaviors, they are both types of eating disorders. Both disorders have negative effects on physical and mental health.

This article contributes to the work on this topic in several ways. This article conducts a survey among female athletes at 30 universities across all three division levels. Collecting data from many institutions allows for more generalizability of results and avoids institution-specific factors driving the results. And by focusing the survey on female collegiate track and field athletes, enough observations are generated to run regression analyses. To the best of our knowledge, this is the first study to simultaneously examine the competition level, team pressures, and socially related pressures faced by female college athletes at many universities. Controlling for a wide range of factors and collecting data from different levels of university competition allows for parsing out the effects of the varied factors. Results find that after controlling for the different pressures, the division level of competition is not a significant factor with regard to eating disorders for the sample. Social pressures were found to impact dieting while it was teammate- and coach-related pressures that increased vomiting behavior. These results show it is necessary to gather data on both different types of pressures and disorders, as the effects are not the same. Current interventions and informational programs are not sufficient to counteract the pressures being felt by college athletes.

The next section will review some of the relevant literature on the topic. This is followed by a description of the survey design and variables. The logit methodology and results are then discussed. This article concludes with the implications of this article for both the literature and policy.

Review of the Literature

There are several studies which examine whether athletes are more at risk for eating disorders. Research has found that athletes were more likely to possess particular behavioral and psychological characteristics linked to eating disorders when compared to non-athletes (Blinde & Taub, 1992). Alexander (1995) focused on female athletes in sororities and finds that women who participate in dance or athletics may be at more risk for eating disorders. Martinsen and Sundgot-Borgen (2013) examined the prevalence of eating disorders among adolescent elite athletes and determined it higher than in non-athletes of the same age. Other scholars have also concluded that female collegiate athletes are more at risk for developing eating disorders (Berry & Howe, 2000).

There are particular subgroups of athletes who may be more prone to develop eating disorders than other athletes. Specifically, athletes who participate in sports that emphasize leanness are at a greater risk of developing eating disorders (Jevne & Stoutjesdyk, 1993; Martinsen et al., 2010; Picard, 1999; Sundgot-Borgen, 1994; Thompson & Trattner-Sherman, 1999). Eating disorders may be caused by the stereotypes in “lean” sports (Koenigsberger, 2016). For track athletes, body stereotypes are based on the type of runner. Sprinters are expected to have a body that is muscular and strong, whereas long distance runners are expected to be strong but very thin. Some research indicates that factors such as the attire required to be worn in competition can increase the risk of disordered eating. When athletes must wear tight and revealing attire, such as the uniforms for gymnastics or distance running, they may strive to be as thin as possible to feel acceptable in the uniforms they are required to wear (Brownell et al., 1992; Giordano, 2010).

When female athletes are pushed to higher levels of elite competition, eating disorders can become a larger problem. A study of female distance runners found that it is the better athletes who are more likely to exhibit the physical and psychological signs of anorexia nervosa (Noakes & Weight, 1987). This is consistent with Picard’s (1999) work which concludes that athletes at higher levels of competition displayed more signs of and had a higher risk for developing eating problems. A meta-analysis of female college athletes concluded elite athletes, especially in sports emphasizing thinness, were more at risk of eating disorders (Smolak et al., 2000).

Consistent with the work on elite competition, the pressure to perform can lead to eating disorders. A qualitative, interview-based study by Arthur-Cameselle et al. (2017) identified performance pressures impacting athletes’ eating disorders and for non-athletes family and other social issues having an impact on eating disorders. In a qualitative study of female collegiate athletes, Arthur-Cameselle and Quatromoni (2010) found that comments by coaches and performance pressure and teammate peer effects impact eating disorder behaviors. This is consistent with a survey of female athletes at Campbell University which found more eating disorders in lean sports and that pressures impact the prevalence of eating disorders (Wells et al., 2015).

Social pressures, unrelated to athletic performance, can also have an impact on eating disorders among college students. Directly related to social pressures, a survey of 246 college women at a small liberal arts college, found that sorority membership increased the prevalence of eating disorders (Basow et al., 2007). However, Averett et al. (2017) did not find a relationship between sorority membership and vomiting or pills/dieting among college students. While not focused on college students, Costa-Font and Jofre-Bonet (2008) found that peers can increase anorexia and feelings of being too fat.

There are several other factors which might affect college students developing eating disorders. Fragkos and Frangos (2013) found the risk occurred more with females, students with divorced parents, students who lived alone, students married or seeking a romantic relationship, and students who were attending a post-secondary school or university and receiving below average level grades. Betz and Mintz (1988) studied eating disordered behaviors among a sample of undergraduate women and found that only 33% of participants reported what could be considered normal eating habits. Some of the risk factors correlated with this include low self-esteem, negative body image, strong tendency to endorse sociocultural beliefs regarding female thinness desirability, obsessive weight and appearance thoughts, and interference of weight and appearance issues with other parts of life. The desire for perfection and high achievement can also lead to eating disorders (Kern & Hastings, 1995; Laliberte et al., 1999; Woodside et al., 2002).

The evidence on athletes and eating disorders is not undeviating in its results. There has been research which finds athletes to not be at a higher risk for developing eating disorders than the general population (Blessing et al., 1990; Jevne & Stoutjesdyk, 1993; Kirk et al., 2001). It could even be that female student athletes have high levels of self-concept and are at a lower risk to develop an eating disorder (Hardin et al., 2014). Also, there may be situations in which female athletic participation is a risk factor for certain elements of eating problems, but at the same time, under other circumstances, participation in sports may protect women from developing issues with eating (Smolak et al., 2000).

Testable Hypotheses

Distance runners are more likely to exhibit eating disorder behaviors. This is consistent with the literature which shows athletes from leaner sports exhibiting more eating disorders.

Athletes competing at a more competitive division level are more likely to have eating disorders. This is predicted by previous research showing increased eating disorders among elite athletes.

Increased pressure will result in more eating disorders. This hypothesis is consistent with the literature on both social and performance pressures impacting eating disorders.

Being overweight as a child is more likely to result in eating disorders. Eating disorders observed in college may be the result of previous circumstances.

Survey and Data Set

The target demographic for the survey was female college students who are members of track and field teams. This comprises any type of track and field athlete including long-distance runners, mid-distance runners, sprinters, throwers, jumpers, and pole vaulters. The survey was created using Qualtrics Survey Software distributed via a link to several dozen collegiate women’s teams of varying sizes and division levels. The survey was distributed in February 2019.

The survey consisted of 40 questions and took approximately 5 min to complete. A number of questions asked in the survey were about eating disorder behaviors. The survey also contained questions on competitive division level and type of athlete. A significant number of questions were asked about different types of pressure on the respondent to be thin. These questions concern both personal and professional pressures to be thin. In addition, the survey gathers information on the general behavior and demographics of the respondents. A copy of the survey is available in the online supplemental material.

The survey yielded 254 fully completed survey responses from 30 institutions. Long-distance runners, mid-distance runners, and sprinters make up almost three quarters of the sample, while athletes who compete in field events make up the remaining quarter. This proportion of track to field athletes is similar to the spread of athletes on a typical collegiate track and field team. Division III athletes make up almost half of the sample (48%), but Division I and Division II athletes are still well-represented at 31% and 21%, respectively. This is close to the actual distribution of Division I, II, and III colleges across the country of 32%, 28%, and 40%, respectively (National Collegiate Athletic Association [NCAA], 2019).

Variables and Methodology

The analysis contains several dependent and several independent variables. The survey yields two eating disorder behaviors with enough observations to be viable in an analysis. The first dependent variable is vomiting or using laxatives to lose weight. The second dependent variable is dieting to lose weight. There is a third dependent variable representing whether the individual has ever been diagnosed (Diag) with an eating disorder. There are several independent variables related to the hypotheses. The first hypothesis relates to the impact of being a distance runner. The variable Distance takes on the value of 1 if the respondent is a mid- or long-distance runner and zero otherwise. Sprinters compete in the 100, 200, and 400 m events. Middle distance is considered 800 and 1,500 m. Long-distance running is 3,000, 5,000, and 10,000 m. Distance runners are considered any athlete running the 800 m or longer race. The sign on this variable is expected to be positive.

To test the second hypothesis, the division level of the college is included. Collegiate sports teams in the United States are separated into three divisions. These divisions represent different levels of competition. Division I represents the highest level of competitive play. Athletes playing on Division I sports teams tend to receive the highest number of scholarships and funding but also have to devote a very substantial amount of time to their sport. It is common for Division I sports athletics to be televised, become a source of university revenue, and can even lead to professional playing opportunities. The increased visibility and financial implications of Division I athletics result in a much larger strain put on athletes to perform well. Several states are seeking to follow the example of California which recently passed a law allowing collegiate athletes to profit off their name and likeness (Berkowitz, 2019). Division I athletics has become, in some cases, a semi-pro league. The hypercompetitive nature of sports teams falls from Division I to Division II and again in the move to Division III. Thus, the sign on division is expected to be negative as a higher division level represents a less competitive division of play.

For the third hypothesis, the total pressure on the respondent to be thin (pressures) is represented on a scale from 0 to 5, with higher values denoting more pressure. This variable is then broken down into two component measures: team pressures and social pressures. Team pressures measures pressure to be thin from coaches and teammates. Each source is equal to one, so the variable ranges from zero to two. Social pressures comprise pressure to be thin from family, friends, and significant others. Similarly, each source is equal to one with this ranging from zero to three. Another measure of social pressure is the variable sorority which is a dichotomous variable representing sorority membership. Pressures to be thin are expected to have a positive relationship with eating disorder behaviors. The fourth hypothesis is tested through a dichotomous variable measuring whether they were viewed as overweight as a child.

There are several variables included to control for other individual behaviors and attributes. There is a variable measuring access to a team nutritionist to control for access to information. To check for if nutritionists have a differential impact by division level, an interaction variable for nutritionist × division was created. There are five variables controlling for individual behaviors which might impact eating disorders. Self-diagnosed stress levels are measured on a 1 to 5 basis with five representing most stressed. Similarly, self-reported involvement in extracurricular activities is also on a 1 to 5 scale with 5 representing the heaviest involvement. There are dichotomous variables representing whether a person considers themselves to be a perfectionist and if they get enough sleep at night. To test for whether a tendency toward perfectionism increases the impact of pressures on behavior, an interaction variable for pressures × perfect was created. A student’s grade point average is also controlled for through the variable GPA. This variable has five different responses with values of 1 to 5 representing a GPA of 4.0 to 3.5, 3.49 to 3.0, 2.99 to 2.5, 2.49 to 2.0, and less than 2.0, respectively. There is a variable to control for race, with 1 representing White and 0 non-White.

Summary statistics of the variables used in the analyses are contained in Table 1 in the following. To test for multicollinearity, a variance inflation factor (VIF) test was run on the independent variables. All VIF results were under 2, which suggests that multicollinearity is not a significant problem in the data.

Table 1.

Descriptive Statistics.

Variable	Observations	Mean	SD	Minimum	Maximum
Dependent variables
Vomiting (V)	254	0.13	0.34	0	1
Dieting (D)	254	0.43	0.50	0	1
Diagnosed (Diag)	254	0.10	0.30	0	1
Independent variables of interest
Distance runner (R)	254	0.48	0.50	0	1
Division level (L)	254	2.16	0.88	1	3
Pressures (P)	254	1.07	1.32	0	5
Team pressures (P^T)	254	0.38	0.66	0	2
Social pressures (P^S)	254	0.69	0.87	0	3
Sorority (S)	254	0.07	0.26	0	1
Overweight as Child (O)	254	0.15	0.36	0	1
Control variables (C)
Nutritionist	254	0.48	0.50	0	1
Stress level	254	3.44	0.76	1	5
Involvement	254	2.14	0.93	1	5
Perfect	254	0.67	0.47	0	1
Enough sleep	254	0.70	0.46	0	1
GPA category	254	1.57	0.73	1	4
Race dummy	254	0.78	0.42	0	1

Note. GPA = grade point average.

Given the dichotomous nature of the dependent variables, a logit approach is utilized. The logit equations are as follows:

V_{i} = α_{0} + α_{1} R_{i} + α_{2} L_{i} + α_{3} P_{i} + α_{4} S_{i} + α_{5} O_{i} + α_{6} C_{i} + ε_{i}

(1)

V_{i} = α_{0} + α_{1} R_{i} + α_{2} L_{i} + α_{3} P_{i}^{T} + α_{4} P_{i}^{S} + α_{5} S_{i} + α_{6} O_{i} + α_{7} C_{i} + ε_{i}

(2)

D_{i} = β_{0} + β_{1} R_{i} + β_{2} L_{i} + β_{3} P_{i} + β_{45} S_{i} + β_{5} O_{i} + β_{6} C_{i} + ε_{i}

(3)

D_{i} = β_{0} + α β_{1} R_{i} + β_{2} L_{i} + β_{3} P_{i}^{T} + β_{4} P_{i}^{S} + β_{5} S_{i} + β_{6} O_{i} + β_{7} C_{i} + ε_{i}

(4)

The variables are as described above with $ε_{i}$ representing the error term. Each of the dependent variables is tested in two different logit equations. The difference between the two logits is whether pressures are represented by the combined total pressures variable $(P_{i})$ or by the team pressures $(P_{i}^{T})$ and social pressures variables $(P_{i}^{S})$ .

Results

The results of the vomiting and dieting logits are in Tables 2 and 3, respectively. The tables display odds ratios and t-statistics and indicate levels of significance. As logit coefficients are not true magnitudes, odds ratios are instead displayed. Odds ratios are always positive, and it is their relation to one which determines the direction of the variable. An odds ratio greater than one is interpreted as the independent variable increasing the likelihood (positive effect) and odds ratios less than one decreasing the likelihood (negative effect). As an example of interpreting magnitudes, an odds ratio of 1.10 would mean that a one unit change in the independent variable results in a 10% higher likelihood of the dependent variable being equal to 1 versus 0.

Table 2.

Logistic Regressions on Vomiting Behavior.

Variables	Model 1	Model 2	Model 3
Distance	2.93 (2.28)**	2.77 (2.15)**	2.93 (2.23)**
Division	1.19 (0.62)	1.22 (0.70)	1.47 (0.75)
Pressures (total)	1.61 (3.34)***
Team Pressures		2.29 (2.56)**	3.78 (2.98)***
Social Pressures		1.25 (0.84)	2.13 (1.88)*
Overweight as Child	4.79 (3.25)***	4.80 (3.26)***	4.95 (3.30)***
Sorority	2.13 (1.01)	2.05 (0.96)	2.20 (1.00)
Nutritionist	1.05 (0.11)	1.03 (0.06)	1.89 (0.41)
Stress	1.23 (0.75)	1.26 (0.84)	1.30 (0.95)
Involvement	1.01 (0.06)	1.05 (0.20)	1.06 (0.24)
Perfect	2.42 (1.65)*	2.59 (1.75)*	9.72 (2.18)**
Sleep	0.81 (−0.45)	0.82 (−0.43)	0.72 (−0.71)
GPA	1.2 (0.57)	1.26 (0.72)	1.26 (0.72)
Race	0.88 (−0.21)	0.82 (−0.32)	0.77 (−0.41)
Pressures × Perfect			0.52 (−1.76)*
Nutritionist × Division			0.78 (−0.38)
Prob >χ²	0.00	0.00	0.00
Number of observations	254	254	254

Note. Odds ratios listed with z-statistics in parentheses. GPA = grade point average.

Denote significance at the 10% level.

Denote significance at the 5% level.

***

Denote significance at the 1% level.

Table 3.

Logistic Regressions on Dieting Behavior.

Variables	Model 1	Model 2	Model 3
Distance	1.84 (1.90)*	1.85 (1.91)*	1.86 (1.92)**
Division	0.74 (−1.54)	0.74 (−1.56)	0.70 (−1.13)
Pressures (total)	1.59 (3.80)***
Team Pressures		1.47 (−1.56)	1.68 (1.72)*
Social Pressures		1.67 (2.67)***	1.95 (2.39)**
Overweight as Child	7.70 (4.48)***	7.71 (4.48)***	7.73 (4.47)***
Sorority	5.46 (2.89)***	5.51 (2.90)***	5.44 (2.86)***
Nutritionist	0.79 (−0.67)	0.79 (−0.68)	0.67 (−0.42)
Stress	1.12 (0.60)	1.12 (0.59)	1.12 (0.58)
Involvement	0.91 (−0.58)	0.91 (−0.62)	0.89 (−0.69)
Perfect	0.81 (−0.62)	0.81 (−0.65)	1.00 (0.00)
Sleep	0.86 (−0.43)	0.86 (−0.45)	0.86 (−0.44)
GPA	0.67 (−1.74)*	0.67 (−1.76)*	0.67 (−1.74)*
Race	0.85 (−0.42)	0.86 (−0.40)	0.86 (−0.41)
Pressures × Perfect			0.81 (−0.79)
Nutritionist × Division			1.08 (0.20)
Prob >χ²	0.00	0.00	0.00
Number of observations	254	254	254

Note. Odds ratios listed with z-statistics in parentheses. GPA = grade point average.

Denote significance at the 10% level.

Denote significance at the 5% level.

***

Denote significance at the 1% level.

The empirical work produced several interesting results. The first hypothesis regarding distance runners was validated. Distance was found to be positive and significant with respect to both eating disorders, with the effect being stronger on vomiting behavior than on dieting. This effect was found for the distance variable which combined both mid- and long-distance runners. This is true both in terms of significance level and magnitude, 2.93 for vomiting, and 1.84 for dieting. These are significantly higher than the estimates found by Martinsen et al. (2010) and Picard (1999) who found distance runners were 11% and 60% more likely to exhibit eating disorders, respectively. As a robustness check, it was also tested using only long-distance runners and results were consistent. Surprisingly, division level was not found to be significant in either regression. This is important as much of the prior research which found the relationship between competitive athletes and eating disorders did not use a sample of universities across all division levels.

Overall pressures were significant with respect to increasing both vomiting and dieting behavior at roughly the same magnitudes. This provides estimates to a literature which has been largely qualitative in nature (Arthur-Cameselle et al., 2017; Arthur-Cameselle & Quatromoni, 2010). There were interesting differences between the vomiting and dieting regressions with regard to the team and social pressure variables. There appears to be a significant difference between performance and socially motivated pressures. Team pressures (from coaches and teammates) had a significant positive impact on vomiting but generally not on dieting. Dieting was more driven more strongly by social pressures. There is weak evidence through the interaction term that the impact of pressures on vomiting may be somewhat mitigated by perfectionism. While not expected, the effect is weak. Neither of the interaction variables were significant with respect to dieting.

Both the social pressures and sorority variables were positive and significant with respect to dieting. The sorority variable had one of the highest magnitude effects on dieting, with sorority membership related with individuals being more than 5 times as likely to engage in this behavior. This is a much stronger effect of sorority membership than found by Basow et al. (2007), although their study was not focused on athletes. However, one cannot be certain how much of this effect may be from the pressure exerted by sorority membership versus a selection effect of who is more likely to join a sorority. Overall, these results suggest the importance of testing different types of eating disorders and pressures. Testing “pressures” as a single variable may not be capturing some of the nuances of how different types of pressures impact people’s eating disorder behaviors.

The fourth hypothesis was supported by the analyses. Being considered overweight as a child had a positive and significant relationship with both vomiting and dieting behavior. In terms of magnitude of the effect, overweight is the most significant variable in both the vomiting and dieting behavior logits. Being considered overweight as a child is related to an almost 5 times higher likelihood of vomiting behavior and more than 7 times more likely to engage in dieting behaviors. There were a couple of notable results from the control variables. Having a high GPA seemed to be a mitigating factor, making dieting less likely. The drive for perfection was positively and weakly significantly related to vomiting behavior. The other control variables were insignificant which is relevant as well. In particular, having a team nutritionist did not affect eating disorder behavior, so lack of access to information is not an important factor. Other individual-level factors such as self-reported stress levels, amount of sleep, extracurricular involvements, and race were not significant.

Behavior Versus Diagnosis

One of the issues with eating disorders can be a disconnect between behaviors and diagnosis (Pritts & Susman, 2003). Individuals can exhibit eating disorders without being formally diagnosed. This can be problematic as people who get diagnosed are more likely to receive treatment. For this reason, it can be instructive to know whether the same factors drive both the likelihood of diagnosis and behaviors. The survey contains a dichotomous variable for whether the athlete has been formally diagnosed with an eating disorder. The logit regression is as follows, with the results in Table 4.

D i a g_{i} = γ_{0} + γ_{1} R_{i} + γ_{2} L_{i} + γ_{3} P_{i}^{T} + γ_{4} P_{i}^{S} + γ_{5} S_{i} + γ_{6} O_{i} + γ_{7} C_{i} + ε_{i}

(5)

Table 4.

Logistic Regressions on Being Diagnosed.

Variables	Model 1	Model 2	Model 3
Distance	55.42 (4.23)**	56.37 (4.23)**	106.69 (4.28)***
Division	0.61 (−1.43)	0.61 (−1.44)	2.84 (1.29)
Pressures (total)	1.29 (1.46)
Team Pressures		1.20 (0.47)	1.56 (0.95)
Social Pressures		1.37 (1.02)	2.03 (1.57)
Overweight as Child	11.96 (3.73)**	12.15 (3.72)**	20.37 (3.76)***
Sorority	0.92 (−0.06)	0.95 (−0.04)	2.87 (0.80)
Nutritionist	0.78 (−0.41)	0.79 (−0.40)	99.28 (1.92)**
Stress	1.15 (0.45)	1.14 (0.41)	1.14 (0.40)
Involvement	0.99 (−0.03)	0.98 (−0.04)	1.02 (0.06)
Perfect	1.09 (0.17)	1.08 (0.14)	1.91 (0.73)
Sleep	0.66 (−0.76)	0.65 (−0.77)	0.57 (−0.99)
GPA	1.39 (0.91)	1.36 (0.84)	1.38 (0.85)
Race	0.38 (−1.40)	0.38 (−1.39)	0.35 (−1.43)
Pressures × Perfect			0.62 (−1.15)
Nutritionist × Division			0.11 (−2.13)**
Prob >χ²	0.00	0.00	0.00
Number of observations	254	254	254

Note. Odds ratios listed with z-statistics in parentheses. GPA = grade point average.

Denote significance at the 10% level.

Denote significance at the 5% level.

***

Denote significance at the 1% level.

Consistent with the behavior regressions, being a distance runner and overweight as a child are both positively related to being diagnosed with an eating disorder. The magnitudes on distance and overweight are far more powerful than any other variables in the diagnosis logit. None of the pressure variables or the pressure interaction are significant with respect to getting diagnosed with an eating disorder. This is relevant as the pressure variables were highly significant factors in affecting eating disorder behaviors. This suggests that team and social pressures may be leading to individuals exhibiting eating disordered behaviors but they are not getting the diagnosis which they need to help them get treatment. However, having a team nutritionist had a positive effect on the likelihood of getting diagnosed, which suggests that nutritionists are helping. The nutritionist × division variable was negative and significant. This suggests that the positive impact nutritionists are having on the likelihood of getting diagnosed is primarily at the most competitive (Division I) level. As the division level gets less competitive, nutritionists have less impact on diagnosis.

Conclusion

The pressures on female collegiate athletes are intense and complex. The results from this article help to shed more light on these pressures. Female athletes are coming under pressure from teammates and coaches which lead to eating disorders such as vomiting. These athletes are also college students as well. This leads to social pressures on them which can increase dieting behaviors.

Understanding the different mechanisms through which eating disorder pressures operate on women as both athletes and students is important for creating appropriate solutions. The empirical work identifies distance runners as the most susceptible group. And these pressures operate across all competition levels, it is not a problem confined to Division I schools. Having a team nutritionist on hand does not seem to remedy the situation. Another concerning result is the finding that these team and social pressures seem to be increasing eating disorder behaviors but not leading people to getting help through diagnosis.

Informational programs are not enough to address the eating disorders being faced by female college athletes. This is a problem which crosses different size and competition level universities. Smaller colleges and/or division III schools must deal with this problem as well. Dealing with this problem will require more intensive programs. Eating disorder screenings need to become a more regular part of athlete’s routines. Coaches can actively work to reduce the pressures they place on their athletes regarding thinness but also need to try and change team culture as peer pressures can be very harmful. And counselors are necessary to help address the “off the track” pressures encountered by female college athletes that may not be well-suited to coaches. Students are coming into college having prior experience with eating disorders and they encounter social pressures separate from being an athlete. Programs need to focus on both the athlete and the student while recognizing that these pressures may lead to different disorders.

This article suggests several additional avenues for future research. First, it would be interesting to collect a sample of male and female collegiate track athletes to determine how these results vary by gender. Also, these results are not necessarily generalizable to other collegiate sports so more testing is required. It would also be very useful to the literature if researchers developed a panel data set so eating disordered behavior could be tracked throughout athlete’s careers.

Supplemental Material

Online_Appendix – Supplemental material for College Athletes Under Pressure: Eating Disorders Among Female Track and Field Athletes

Supplemental material, Online_Appendix for College Athletes Under Pressure: Eating Disorders Among Female Track and Field Athletes by Michael A. Quinn and Stephanie Robinson in The American Economist

Footnotes

Authors’ Note

Stephanie Robinson is now affiliated with Deloitte Consulting, Boston, MA.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Michael A. Quinn

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Michael A. Quinn is an economics professor at Bentley University. He received his undergraduate economics degree from Fairfield University and his PhD in economics from The University of North Carolina at Chapel Hill. His research areas include economic development and international economics.

Stephanie Robinson received an undergraduate degree in Economics-Finance from Bentley University where she also competed as a varsity track athlete. Stephanie is currently working as a Risk and Financial Advisory consultant at Deloitte.

References

Alexander

L. A.

(1995). The prevalence of eating disorders and eating disordered behaviors in sororities (Master’s thesis 1911–February 2014, 2292). https://scholarworks.umass.edu/theses/2292

Arthur-Cameselle

J. N.

Quatromoni

P. A.

(2010). Factors related to the onset of eating disorders reported by female collegiate athletes. The Sport Psychologist, 25, 1–17.

Arthur-Cameselle

J. N.

Sossin

Quatromoni

P. A.

(2017). A qualitative analysis of factors related to eating disorder onset in female collegiate athletes and non-athletes. Eating Disorders, 25(3), 199–215.

Averett

Terrizzi

Wang

(2017). The effect of sorority membership on eating disorders, body weight, and disordered-eating behaviors. Health Economics, 26(7), 875–891.

Basow

S. A.

Foran

K. A.

Bookwala

(2007). Body objectification, social pressure, and disordered eating behavior in college women: The role of sorority membership. Psychology of Women Quarterly, 31(4), 394–400.

Berkowitz

(2019, September 30). California Governor Signs Bill that Makes it Easier for College Athletes to Profit from Name, Likeness. USA Today. https://www.npr.org/2019/09/30/765700141/california-governor-signs-bill-allowing-college-athletes-to-profit-from-endorsem

Berry

T. R.

Howe

B. L.

(2000). Risk factors for disordered eating in female university athletes. Journal of Sport Behavior, 23(3), 207–218.

Betz

N. E.

Mintz

L. B.

(1988). Prevalence and correlates of eating disordered behaviors among undergraduate women. Journal of Counseling Psychology, 33(4), 463–471.

Blessing

D. L.

Stanton

A. L.

Warren

B. J.

(1990). Disordered eating patterns in competitive female athletes. International Journal of Eating Disorders, 9(5), 565–569.

10.

Blinde

E. M.

Taub

D. E.

(1992). Eating disorders among adolescent female athletes: Influence of athletic participation and sport team membership. Adolescence, 27(108), 833–848.

11.

Brownell

K. D.

Rodin

Wilmore

J. H.

(1992). Eating, body weight, and performance in athletes: Disorders of modern society. Lea and Febiger.

12.

Costa-Font

Jofre-Bonet

(2008). Body image and food disorders: Evidence from a sample of European women (CESifo working paper no. 2412). https://ideas.repec.org/p/ces/ceswps/_2412.html

13.

Eisenberg

Nicklett

Roeder

Kirz

(2011). Eating disorders symptoms among college students: Prevalence, persistence, correlates, and treatment-seeking. Journal of American College Health, 59(8), 700–707.

14.

Fragkos

K. C.

Frangos

C. C.

(2013). Assessing eating disorder risk: The pivotal role of achievement anxiety, depression and female gender in non-clinical samples. Nutrients, 5(3), 811–828.

15.

Gilbert

A. A.

Shaw

S. M.

Notar

M. K.

(2000). The impact of eating disorders on family relationships. Eating Disorders, 8(4), 331–345.

16.

Giordano

(2010). Exercise and eating disorders: An ethical and legal analysis. Routledge.

17.

Hardin

Hoppe

McLester

C. N.

(2014). Susceptibility to eating disorders among collegiate female student-athletes. Journal of Athletic Training, 49(3), 406–410.

18.

Hudson

J. I.

Hiripi

Pope

H. G.

Kessler

R. C.

(2007). The prevalence and correlates of eating disorders in the national comorbidity survey replication. Biological Psychiatry, 61(3), 348–358.

19.

Jevne

Stoutjesdyk

(1993). Eating disorders among high performance athletes. Journal of Youth and Adolescence, 22(3), 271–282.

20.

Kern

J. M.

Hastings

(1995). Differential family environments of bulimics and victims of childhood sexual abuse: Achievement orientation. Compulsive Eating and Depression, 51(4), 499–506.

21.

Kirk

Singh

Getz

(2001). Risk of eating disorders among female college athletes and nonathletes. Journal of College Counseling, 4:, 122–132.

22.

Koenigsberger

A. A.

(2016). Fair play: On the unjust stereotypes in sports practice. Revista de Filosofía, Ética y Derecho del Deporte, 1(7), 23–47.

23.

Laliberte

Boland

F. J.

Leichner

(1999). Family climates: Family factors specific to disturbed eating and bulimia nervosa. Journal of Clinical Psychology, 55(4), 1021–1040.

24.

Le Grange

Swanson

S. A.

Crow

S. J.

Merikangas

K. R

. (2012). Eating disorder not otherwise specified presentation in the US population. International Journal of Eating Disorders, 45(5), 711–718.

25.

Martinsen

Bratland-Sanda

Eriksson

A. K.

Sundgot-Borgen

(2010). Dieting to win or to be thin? A study of dieting and disordered eating among adolescent elite athletes and non-athlete controls. British Journal of Sports Medicine, 44(1), 70–76.

26.

Martinsen

Sundgot-Borgen

(2013). Higher prevalence of eating disorders among adolescent elite athletes than controls. Medicine and Science in Sports and Exercise, 45(6), 1188–1197.

27.

Muñoz

Quintana

J. M.

Las Hayas

Aguirre

Padierna

Gonzalez-Torres

M. A.

(2009). Assessment of the impact of eating disorders on quality of life using the disease-specific, Health-Related Quality of Life for Eating Disorders (HeRQoLED) Questionnaire. Quality of Life Research, 18(9), 1137–1146.

28.

National Center on Addiction and Substance Abuse. (2003). Food for thought: Substance abuse and eating disorders. Columbia University.

29.

National Collegiate Athletic Association. (2019). Our three divisions. http://www.ncaa.org/about/resources/media-center/ncaa-101/our-three-divisions

30.

Noakes

T. D.

Weight

L. M.

(1987). Is running an analog of anorexia? A survey of the incidence of eating disorders in female distance runners. Medicine and Science in Sports and Exercise, 19(3), 213–217.

31.

Picard

C. L.

(1999). The level of competition as a factor for the development of eating disorders in female collegiate athletes. Journal of Youth and Adolescence, 28(5), 583–594.

32.

Pritts

S. D.

Susman

(2003). Diagnosis of eating disorders in primary care. American Family Physician, 67(2), 297–304.

33.

Smink

F. E.

van Hoeken

Hoek

H. W.

(2012). Epidemiology of eating disorders: Incidence, prevalence and mortality rates. Current Psychiatry Reports, 14(4), 406–414.

34.

Smolak

Murnen

S. K.

Ruble

A. E.

(2000). Female athletes and eating problems: A meta-analysis. International Journal of Eating Disorders, 27(4), 371–388.

35.

Sundgot-Borgen

(1994). Eating disorders in female athletes. Sports Medicine, 17(3), 176–188.

36.

Thompson

R. A.

Trattner-Sherman

(1999). Athletes, athletic performance, and eating disorders: Healthier alternatives. Journal of Social Issues, 55(2), 317–337.

37.

Volpe

Tortorella

Manchia

Monteleone

A. M.

Albert

U. P.

Monteleone , . (2016). Eating disorders: What age at onset? Psychiatry Research, 238, 225–227.

38.

Wells

E. K.

Chin

A. D.

Tackle

J. A.

Bunn

J. A.

(2015). Risk of disordered eating among division I female college athletes. International Journal of Exercise Science, 8(3), 256–264.

39.

Woodside

D. B.

Bulik

C. M.

Halmi

K. A.

Fichter

M. M.

Kaplan

Berrettini

W. H.

Strober

Treasure

Lilenfeld

Klump

Kaye

W. H.

(2002). Personality, perfectionism, and attitudes toward eating in parents of individuals with eating disorders. International Journal of Eating Disorders, 31(3), 290–299.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB